This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
IntroductionThis technical guide addresses the preparation and gas chromatographic (GC) analysis ofsemivolatile organic compounds such as those listed in US Environmental Protection Agency(EPA) Methods 8270, 525, and 625; and polycyclic aromatic hydrocarbons (PAHs) such asthose listed in US EPA Methods 610 and 8100. These analyses are some of the most commontests performed by environmental laboratories, yet there are many analytical challenges ofwhich the analyst needs to be aware. For example, the samples often are highly contaminatedwith non-target compounds (e.g., hydrocarbons) and quality assurance/control (QA/QC) ofthe methods is rigorous. There are several procedures and techniques that can be employed,however, to make these analyses simpler to perform. Review this guide to learn thesetechniques and to troubleshoot analytical problems associated with the methods.
The compounds addressed in this guide are listed in Table I, but many additional compoundsare also amenable to these semivolatile methods. Table I includes the compounds cited in theUS EPA Methods, as well as some other compounds typically analyzed in environmentalsamples.
ExtractionThe compounds listed in Table I may be difficult to extract because they fall into differentchemical classes (i.e., acidic, basic, neutral, halogenated, oxygenated, polar, non-polar, low-boiling, and high-boiling compounds). Therefore, the extraction method will need to solvate awide variety of compounds. It also must recover the analytes of interest while removing theinterfering non-target contaminants. This limits the choices of cleanup options. A number ofsample extraction methods can be applied to these compounds, but only the most commonwill be addressed in this guide.
Liquid SamplesFor liquid samples, either separatory funnel extraction (US EPA Method 3510) or automatedliquid-liquid extraction (US EPA Method 3520) may be used. Separatory funnel extraction isfaster and less expensive to set up than the other methods, but it requires continuous operatorattention. Automated liquid-liquid extractors run unattended, but are more expensive and, ifanalyte recovery is lower than allowed, re-extraction by separatory funnel may be required.Alternatively, if the sample forms an emulsion to the degree that acceptable solvent recoveryis not possible using a separatory funnel, then some samples will require automated liquid-liquid extraction. Solid phase extraction (US EPA Method 3535) also is an option for aqueoussamples.
For separatory funnel extraction, measure up to 1L of water into a 2L separatory funnel andadjust the pH to >11 using 10M NaOH; be careful not to add too much base. Then extract thesample by adding 60mL of dichloromethane and shaking for two minutes. It is critical toshake all samples consistently or variations in extraction efficiency will be observed. The bestway to ensure consistency is to use a mechanical separatory funnel shaker, as there often isconsiderable variation with manual extractions. Allow the dichloromethane layer to settle tothe bottom of the funnel and then decant that layer into a collection vessel (i.e., a KurdenaDanish [KD] concentrator, or a Turbo vap or Rapid vap® container if using automatedconcentrators). This extraction step is repeated twice more to get quantitative recovery of allanalytes. Collect all three extractions into the same collection vessel and label as base/neutral.
Then adjust the water sample to a pH of slightly less than 2 using sulfuric acid (1:1, v/v).Avoid over-acidification because it can result in an acidic extract. Repeat extraction procedureon the water sample as described above, collecting extracts in a separate collection vessel andlabeling it as acid fraction.
It is critical to remove water from the dichloromethane before you concentrate the extract tofinal volume. Dichloromethane can hold approximately 11mL of water per liter ofdichloromethane. If this water remains in the extract, it will partition out when the volume isreduced. This will result in the dichloromethane boiling off first, leaving water in thecollection vessel, and the formation of a two-layer extract. The analyte recoveries will belower than desired, and the presence of water will interfere with the GC analysis.
Restek’s Customer CommitmentPlus 1™ Service means we will surpass
your expectations every time you contactus! You’ll get Plus 1™ service when youask our experienced Technical ServiceTeam to help solve a difficult analytical
problem. Our helpful, efficient CustomerService Team provides Plus 1™ service
even when you place a late-day order. Keepreaching for Restek products and service,
and we will provide you with Plus 1™
quality and attention.
3
RestekTip
How to Bake Sodium Sulfate
To bake the sodium sulfate, spread itinto a glass pie plate no more than1" thick and place into a mufflefurnace at 400°C for a minimum oftwo hours. After this time, thesodium sulfate should be placed intoa glass container while still hot and
-lined cap toprevent the material fromreadsorbing contaminants from theatmosphere.
Table I. Semivolatile organic compounds listed in US EPA Methods 8270, 525, and 625.
The optimum way to remove the water is to decant the dichloromethane through granularsodium sulfate, which is held in a funnel with a high-quality grade filter paper (e.g.,Whatman® 541). Approximately 30g of sodium sulfate are sufficient for most samples. Thisstep must not be skipped. Methods may call for powdered sodium sulfate, but some analytescan be adsorbed onto the smaller particles so it is recommended that only 10-60 mesh orsimilar granular sodium sulfate be used. Also, it is important that this material be contami-nant-free, so it should be purchased as American Chemical Society (ACS) pesticide residue-grade in glass containers or baked in a muffle furnace if purchased in bulk packages whereexposure to plastic is an issue (see Restek Tip). If a muffle furnace is not available, thesodium sulfate can be washed or extracted with dichloromethane prior to use; however thistechnique uses large amounts of solvent.
sealed with a PTFE
4
Automated liquid-liquid extraction can run unattended once the samples are ready and thesolvent is added. This extraction is performed at a single pH. Generally, you will need toadjust the sample pH to 2, but some methods call for adjusting the pH to 4. In any event, it iscritical to not let the pH go below 2 when using a liquid-liquid extractor. If this happens, anacidic extract will form and may cause damage to the GC column. Acidic extracts also willcause low recoveries for the late-eluting internal standards, possibly due to isotope exchange(e.g., perylene-d12).
Automated liquid-liquid extractors are available in two versions—conventional and acceler-ated. The conventional types use 1L of sample and extract using 100 to 500mL ofdichloromethane. These extraction vessels typically are operated for 16 to 24 hours in order toachieve complete extraction. The accelerated extractor uses a hydrophobic membrane toseparate the aqueous from the organic phases, and the extraction time can be reduced by 25 to30% compared to the conventional extractor. However, the membranes are expensive, so it isimportant to analyze the cost versus the number of samples extracted to determine if there is acost benefit to using this accelerated technique.
Finally, solid phase extraction (SPE) also may be used to extract semivolatile organiccompounds from aqueous samples. When using SPE, it is extremely important to follow themanufacturer’s recommendations for product use. There are several manufacturers of C18cartridges and disks, which are the typical media for these compounds. The specific steps toextract these compounds will vary somewhat depending on the manufacturer. One of thebiggest problems with SPE is plugging of the disk or tube with suspended solids, so thismethod only works reliably for drinking water samples. If contamination levels are low andthe samples are free of solids, SPE provides very fast extraction times and low solvent usage.It is used easily for field extractions. And, generally, the disks are preferred for the extractionof 1L sample volumes, but recoveries are not uniform for all of the compounds in Table I. Thecompounds listed in US EPA Method 525.2 exhibit good extraction recoveries using thistechnique. For detailed information on this extraction, request the Applications Note “SPEExtraction for US EPA Method 525.1” (lit. cat.# 59557).
Soil SamplesSoxhlet and ultrasonic extraction are the most common extraction techniques for solid samples;although pressurized fluid, microwave, and supercritical fluid extraction (SFE) also can be used.
Because the soil and biota samples essentially are wet particles, acetone and dichloromethane(1:1) usually are used as the extraction solvents. Acetone is needed to adequately penetrate thesoil particle so that compounds in the particle can be extracted. Several other solvent systems areused for more specialized extractions, but for most applications this combination works well.
All solvents used for extractions must be ACS pesticide-residue grade, and a solvent assayshould be performed to verify purity prior to use. To perform a solvent assay, evaporate 300 to400mL of solvent to a final volume of 1mL and analyze by GC/mass spectrometry (MS). Theresulting chromatogram should have no compounds quantitated above 1/2 the detection limitfor any target compound.
Soxhlet and ultrasonic extraction work well for the semivolatile compounds listed in Table I.Sonication is a faster technique but requires constant operator attention. In both techniques,problems usually are caused by contaminated reagents (especially sodium sulfate) or byinconsistent handling from sample to sample. Sodium sulfate must be treated to remove wateras described in the Restek Tip on page 3, and the sample must be mixed with the sodiumsulfate to achieve a sandy consistency.
Pressurized fluid extraction (US EPA Method 3545A) can be run in an unattended fashionwith multiple samples across a wide sample size range. Extraction vessels with volumes of 1to 100 mL are available. Instruments like the Dionex ASE 200 accommodate wet samplesfrom 1 to 15 grams, and the Dionex ASE 300 will accommodate wet samples from 15 to 50grams. The volume of the cell needed for wet samples is generally twice the gram weight ofthe sample being used. For example, if 30-g wet samples are needed, the 66-mL and 100-mLvessels will be adequate for these extractions. This is necessary because a drying agent such
RestekTip
Clean Glassware
It is important to properly cleanglassware used during sampleextraction. Contaminated glasssurfaces can react with samples andcause breakdown or adsorption ofactive compounds. Verify cleanlinessby running blanks through allglassware.
5
RestekTip
Stabilizing Dichloromethane
Dichloromethane requires a stabilizer toprevent the formation of hydrochloricacid (HCl). Without a stabilizer, HClwill form and injection of acidicdichloromethane will cause inlet linersand columns to become reactive. Thereare two types of stabilizers: stabilizersthat keep HCl from forming, andstabilizers that eliminate HCl uponformation. Methanol is a stabilizer thatprevents HCl from forming; whereascyclohehane, cyclohexene, 2-methylbutene, and amylene scavengethe HCl after its formation.
Dichloromethane used in liquidextractors should contain both types ofstabilizers. Methanol is a betterstabilizer, acting as a free radicalinhibitor, but methanol partitions intothe the water phase. This could leave anunstabilized extract unless a scavengerstabilizer also is used.
as diatomaceous earth is added to the sample prior to being loaded into the extraction vessels.The type of samples being extracted as well as the required method detection limits should beconsidered as part of the evaluation of pressurized fluid extraction.
Microwave extraction (US EPA Method 3546) can be useful for automated extraction as well.This method typically performs the extraction of 12 samples simultaneously, but requires slightlymore operator handling than the pressurized fluid extraction instruments. Microwave extractioninstrumentation is less expensive, but can suffer from the same sample size limitations.
Supercritical fluid extraction (SFE) has been promoted for a number of years as a means of“solventless” extraction for environmental samples. SFE has been added to SW-846 asMethods 3560, 3561, and 3562 but its application is limited. SFE suffers from severe matrix-related variation, requiring modification of its conditions depending on soil type, watercontent, sample size, and type of analytes. Doing so ultimately requires additional samplepreparation prior to the actual extraction. These requirements, added to the high cost of theinstrument, have virtually precluded the use of SFE for environmental sample preparation.
CleanupSample extract cleanup may be the most important step in maintaining long-term instrumentperformance. Many times, when instrument problems arise, they are caused by exposure ofthe injection port and the column to material in the sample extracts other than the targetcompounds. While all contaminants cannot be eliminated, reducing them will minimizeinjection port and column maintenance. Most semivolatile extracts, especially those extractsfrom soil and biota samples, contain high-boiling hydrocarbons and lipids. The difficulty inattempting to remove these compounds using one of the common solid-liquid cleanuptechniques (e.g., Florisil® and silica gel) is that the cleanup technique also removes some ofthe target compounds. In addition, because the analytical method usually calls for thereporting of several tentatively identified compounds (TICs), it is not desirable to clean theextracts of compounds that would normally elute in the range of the target compounds. Forthese reasons, gel permeation chromatography (GPC) is the only universal cleanup techniquefor semivolatile extracts.
Gel Permeation ChromatographyGel permeation chromatography (GPC) is a preparative scale chromatographic method ofseparation based on molecular size. Because the target compounds are similar in molecularsize, they elute as a band of material and are easily separated from lighter and heaviercontaminants. However, GPC systems are expensive and the processing time per sample isbetween 30 to 70 minutes. For these reasons many laboratories choose not to use GPC.However, it is very efficient for removing sulfur, high molecular weight hydrocarbons, andlipids from semivolatile extracts; and may be prudent for soil and biota samples.
Although sulfur can be removed using other techniques such as mercury or activated copperpowder, these procedures, especially copper powder, may degrade some of the target compoundsand will not remove the high-boiling hydrocarbons or lipids. The lipid content of biota extractscan be significant and may overload most SPE clean-up techniques. If a sample extract with ahigh lipid content is injected into the GC, the injection port and front of the column will becomecontaminated quickly. This will result in failure of check standards and the loss of activecompounds such as nitroanilines, nitrophenols, carbazole, and pentachlorophenol (PCP). In spiteof the added expense and time required for GPC, it is the best alternative for extract cleanup.
US EPA Method 3640 details the requirements for GPC cleanup of extracts for semivolatileanalysis. One of the important steps of GPC cleanup is to ensure each day that the instrumentis within its retention time calibration. Although not required by the method, it is goodpractice to run a daily calibration check standard before processing the next batch of samples.If a number of samples have been processed that contain large amounts of contamination, thefront of the GPC column can become reactive. This typically is observed in the loss of 2,4,6-tribromophenol for semivolatile extracts. If the column becomes reactive, injecting blanksmay return the system to control and save the time required to change the column.
6
Obtaining consistent GPC results begins with the extraction and concentration proceduresbecause slight changes in mobile phase and sample solvent composition can result in sometarget compounds being uncollected. Because the typical sample solvent for GPC is puredichloromethane, it is critical that all extracts be reduced to as small a volume as possiblebefore reconstitution in dichloromethane to avoid large amounts of acetone being applied tothe column. Soil and biota samples typically are extracted with a solvent mixture of acetoneand dichloromethane. It is critical that all extracts be reduced to as small a volume as possiblebefore reconstitution in dichloromethane to avoid large amounts of acetone being applied tothe column. Dichloromethane has a lower boiling point than acetone, so it will evaporate firstduring sample concentration, which will leave nearly 100% acetone in the concentrationvessel. If dichloromethane is then added to adjust the extract to volume, there will besignificant amounts of acetone introduced to the GPC column. This will lead to “solventshock” and the formation of a void will be observed at the front of the column. This, in turn,will affect the retention times of the compounds eluting from the GPC column and ultimatelywill result in some target compounds being uncollected. Table II lists the commonsemivolatile compound elution volumes using GPC.
Analysis
Calibration StandardsCalibration standards are purchased as mixtures and usually are divided among three to sevenseparate ampuls due to the cross-reactivity of several compounds. It is important whenmaking the actual working standard that the solution be stored under refrigerated conditions ina Mininert™ vial (Restek cat.# 21050 and 21051) due to the volatility of some of the com-pounds. Failure to properly store the calibration standards will result in evaporative loss of theearly-eluting compounds and the solvent. This will, in effect, concentrate the late-elutingcompounds and cause continuing calibration failure and quantitation errors. Even when storedunder the correct conditions, there still will be degradation of some compounds due to cross-reactivity. This is observed as a loss of the target compound and commonly occurs withbenzidine, 3,3'-dichlorobenzidine, 4-chloroanaline, N-nitrosodiphenylamine, and to a lesserextent with the phenols and other anilines. These standards are stable in the separate ampulssupplied from the manufacturer, but problems arise when all of the compounds are mixedtogether to make the working calibration standard. Therefore, it is important to monitor theresponse of the more active compounds and make fresh mixtures when the calibrationstandards degrade.
Injection Port ConfigurationSeveral of the compounds listed in Table I are prone to breakdown or adsorption on activesurfaces. Typically this will occur in the injection port; therefore, careful attention must begiven to the configuration and maintenance of the injection system.
On-column injection techniques can eliminate breakdown or adsorption in the injectionsystem and improve chromatographic analysis for drinking water extracts or extracts withlittle or no non-volatile residues. However, we do not recommend on-column injections forsoil and biota extracts or extracts that contain large amounts of non-volatile residue, becausethe analytical column can be contaminated quickly.
The preferred injection technique for analyzing highly contaminated extracts is direct injection,but direct injection can cause solvent peak tailing and result in some of the target compoundseluting close to the solvent peak.
To reduce solvent peak tailing, splitless injection is most commonly used for GC/MS analysis ofsemivolatile compounds. There are some drawbacks to splitless injection including molecularweight discrimination, incomplete sample transfer, and reactivity. These problems can beminimized if the technique is properly optimized. Splitless injection requires an injection systemthat is equipped with a solenoid valve controlling the flow to a split vent. The solenoid valve isclosed during the injection process, so the majority of the vaporized sample moves to the front of
RestekTip
Mixing Calibration Standards
When blending several ampuls toproduce a calibration standard, it isimportant that all the compounds arecompletely dissolved in the solvent.This is particularly important withsome of the high molecular weightpolycyclic aromatic hydrocarbons(PAHs) and pesticides that canseparate from solution duringrefrigerated storage. Before openingampuls containing semivolatilecompounds, allow them to warm toroom temperature. Some mixturesmay require sonication to ensurecomplete solubility. Follow themanufacturer’s recommendations forproper handling of the standardmixture. Because some semivolatilecompounds are light sensitive, it isrecommended that calibrationstandards be stored in amber vials.
7
Table II. Mobile phase volumes for elution of semivolatile compounds by GPC.
Compound Elution Volumes (mL)Start End
2-fluorophenol 201 241
phenol-d5 201 249
phenol 209 225
bis(2-chloroethyl)ether 201 225
2-chlorophenol-d4 209 249
2-chlorophenol 209 241
1,3-dichlorobenzene 225 257
1,4-dichlorobenzene-d4 225 257
1,4-dichlorobenzene 225 257
1,2,-dichlorobenzene-d4 225 257
1,2-dichlorobenzene 225 257
2-methylphenol 201 241
2,2'-oxybis(1-chloropropane) 201 225
4-methylphenol 201 241
N-nitroso-di-n-propylamine 186 217
hexachloroethane 233 257
nitrobenzene-d5 209 233
nitrobenzene 209 233
isophorone 193 217
2-nitrophenol 217 233
2,4-dimethylphenol 201 225
bis(2-chloroethoxy)methane 193 225
2,4-dichlorophenol 201 241
1,2,4-trichlorobenzene 225 257
naphthalene-d8 225 249
naphthalene 225 249
4-chloroaniline 217 241
hexachlorobutadiene 225 249
4-chloro-3-methylphenol 201 225
2-methylnaphthalene 225 249
hexachlorocyclopentadiene 225 241
2,4,6-trichlorophenol 201 249
2,4,5-trichlorophenol 201 249
2-flurorbiphenyl 217 241
2-chloronaphthalene 233 249
2-nitroaniline 209 233
dimethylphthalate 193 217
acenaphthylene 225 257
2,6-dinitrotoluene 193 225
Compound Elution Volumes (mL)Start End
acenaphthene-d10 225 249
3-nitroaniline 209 225
acenaphthene 225 249
2,4-dinitrophenol 201 225
dibenzofuran 225 249
4-nitrophenol 201 217
2,4-dinitrotoluene 201 225
diethylphthalate 186 209
fluorene 225 241
4-chlorophenyl-phenylether 217 241
4-nitroaniline 201 225
4,6-dinitro-2-methylphenol 201 225
N-nitrosodiphenylamine 201 233
2,4,6-tribromophenol 217 249
4-bromophenyl-phenylether 217 241
hexachlorobenzene 233 257
pentachlorophenol 209 249
phenanthrene-d10 225 257
phenanthrene 225 257
anthracene 225 257
carbazole 225 257
di-n-butylphthalate 178 201
fluoranthene 225 257
pyrene 225 257
terphenyl-d14 217 233
butylbenzylphthalate 178 201
benzo(a)anthracene 225 257
3,3'-dichlorobenzidine 209 241
chrysene-d12 225 257
chrysene 225 257
bis(2-ethylhexyl)phthalate 162 186
di-n-octylphthalate 162 186
benzo(b)fluoranthene 225 265
benzo(k)fluoranthene 225 265
benzo(a)pyrene 225 265
perylene-d12 249 273
indeno(1,2,3-cd)pyrene 241 273
dibenz(a,h)anthracene 225 257
benzo(g,h,i)perylene 233 273
Standard prepared and loaded as 1/4 acetone, dichloromethane; 5mL sample loop;Column: 70g SX-3 silica size exclusion packing; Guard column: 5g same packing;Flow rate: dichloromethane at 5.3mL/min. at 13 psi; Samples analyzed by GC/MS.
(US EPA 8270 )
the column. After a short time the solenoid valve is opened to allow excess solvent vapor to exitthe split vent. The process of transferring the sample onto the column is relatively slow duringsplitless injection, so the sample must recondense at the front of the column through solvent oranalyte focusing. This is accomplished by having the starting oven temperature 20°C lower thanthe boiling point of the solvent or the first eluting compound.
8
Figure 1. Optimizing splitless hold time.
0 15 30 45 12075 90 10560
late-eluting compounds
hold time (sec.)
20
16
12
8
4
0
com
pone
nt a
rea
(tho
usan
ds)
The time period that the solenoid valve is closed is referred to as the splitless hold-time.The hold-time must be optimized to obtain the best performance from the analytical system.If the solenoid valve is opened too quickly, some of the sample will be lost causing reducedresponse. If the solenoid valve is open too long, the solvent peak will tail. The splitless hold-time will vary depending on column flow rate, injection port geometry, injection porttemperature, and volatility of the analytes. It is impossible to predict the optimum splitlesshold-time without performing some experimentation under the exact conditions of youranalysis.
To optimize the splitless hold time for a particular instrument, prepare a standard that containsboth an early- and a late-eluting compound (e.g., fluorophenol and benzo(g,h,i)perylene).Inject this standard over a range of splitless hold times from 0.1 to 2.0 minutes and plot thedata. An example of this optimization is shown in Figure 1.
In this example the optimum splitless hold time is 60 seconds. This is the point on the graphwhere the response of the late-eluting compound levels off. Holding the solenoid valve closedlonger will not appreciably increase the response of this compound, but will greatly increasethe size of the solvent peak. Because the lower boiling compound will transfer onto thecolumn faster, its response will level off sooner (in this example ~45 seconds). Once this datahas been plotted, it is possible to observe the correct splitless hold-time (once again, the pointat which the response of the late-eluting compounds levels off). The net effect of thisoptimization is to maximize response of late-eluting compounds while minimizing solventtailing.
In addition to optimizing the splitless hold-time, fused-silica wool should be used in theinjection port liner to improve vaporization of higher molecular weight compounds. Whilethere are different theories regarding the placement of fused-silica wool, consistency in theamount of packing and location of the packing is most important. Restek recommends placingthe plug of wool below the point that the syringe needle reaches, but above the inlet of thecolumn. We also recommend using a gooseneck liner to minimize contact between theinjected sample and the bottom of the injection port. This will help improve the response ofthe more reactive compounds such as 2,4-dinitrophenol, PCP, and the nitroanalines. Thegooseneck liner also makes the greatest improvement in response and minimization of endrinbreakdown for US EPA Method 525.
Another technique to minimize molecular weight discrimination is to perform the splitlessinjection under a higher column head pressure. A high inlet pressure is advantageous duringinjection to control the rapidly expanding vapor cloud in the inlet. By using a momentarypressure pulse for the time that the split vent line is closed, the sample vapor cloud iscontrolled and sample backflash into the gas lines entering and exiting the injection port isminimized. The effect of the pressure pulse is to increase the amount of analyte transferred tothe column, especially the late-eluting components. This can lead to stationary phaseoverload, however, so it may be necessary to increase the capacity of the column when usingthis technique (see the Column Selection section for more information).
early-eluting compounds
9
Table III. Recommended installation distances.
Figure 2. The Rtx®-5Sil MS column structure.
Agilent (HP): 5-7mm from tip of ferruleVarian 1075/1077: 5.7cm from back of nutPerkinElmer Autosystem: 4.5 - 5.0cm from back of nutShimadzu 14A: 4.0cm from back of nutShimadzu 17A: 35mm from tip of ferrule
split: 40mm from back of nutsplitless: 64mm from back of nut
Any injection technique can suffer from reactivity (i.e., breakdown) and splitless injection isno exception. The splitless technique has two primary mechanisms for compound reactivity:sample backflash into the gas lines that enter and exit from the injector; and exposure of thesample extract to active sites on the wool, liner, and tip of the column. In general, the same setof compounds break down regardless of which mechanism is occurring: 2,4-dinitrophenol,PCP, 4-nitrophenol, carbazole, and 3-nitroaniline.
Daily maintenance of the injection port will help decrease this problem. Replace the inlet linerand fused silica wool plug, and the septum every day. Weekly, or more often depending on theextract contamination level, replace the inlet seal and remove a short section from the front ofthe column. The length of column removed will vary depending on the level of contaminationin the extracts, generally 6 to 12 inches is adequate. When cutting the column and re-installingit into the injection port, be sure to make a square cut and be consistent with the installationdistance. The installation distance varies by manufacturer. Refer to Table III for a list ofrecommended insertion distances.
Column SelectionDue to the wide variation in functionality, volatility, and polarity of semivolatile compounds,it is not possible to select a column that is highly selective for all of them. As a result, thisanalysis is performed on a general-purpose stationary phase. The Rtx®-5Sil MS column hasthe best combination of low bleed, high inertness, and efficiency for semivolatile applications.(The Rtx®-5MS column also has been successfully used for analysis of semivolatile com-pounds.) The Rtx®-5Sil MS column features a silarylene phenyl/methyl phase that wasdeveloped to provide lower bleed and greater efficiency than other “5-type” phases forimproved separation of the PAHs (Figure 2).
Low-bleed columns are necessary for the more sensitive instruments. For laboratories usingthe Agilent 5973 GC/MS or ion trap MS, column bleed can be a very important issue. Asthese instruments have become more sensitive, the higher-bleed columns produce a larger
signal on the detector and cancause electron multiplier satura-tion. If this occurs, calibrationcurves may show non-linearity athigher concentrations. This issometimes referred to as high-endroll-off, when the signal for agiven concentration is lower thanexpected due to detector satura-tion. (See Quantitation section formore information on high-endroll-off.)
To diagnose column bleed problems, make an injection that allows the bleed to be measuredrelative to the concentration of the analyte in the method. Many data systems will normalizethe display to the largest peak in the analysis. If no compounds are injected, the display willfalsely indicate a high background. The Rtx®-5Sil MS column shows a minimal bleed levelfor a 20ng per component standard (Figure 3).
10
Figure 3. The Rtx®-5Sil MS column exhibits low bleed at 20ng concentration level.
Detector saturation also can be caused by the concentration of the analytes. It was commonpractice on older, less sensitive GC/MS systems to increase the multiplier voltage above the tunevalue to improve sensitivity of low-concentration standards. This technique can lead to problemswith the newer, more sensitive instruments. It is much more likely the higher concentrationcalibration standards will saturate the new GC/MS systems. It may be necessary to reduce themultiplier voltage below the tune value if high-end roll-off is observed. High-end roll-off alsomay be observed when using pressure-pulsing injection techniques to minimize high molecularweight discrimination. If this is observed, you may either increase the stationary phase filmthickness, or increase the column diameter. Alternatively, you may modify the injectionconditions to eliminate the source of the overload.
Column capacity also must be addressed when optimizing the analysis. The typical calibrationrange for many of these methods is 20 to 160ng per compound. This requires a column stationaryphase and diameter that will not overload with a 160ng or larger injection. Because there is a lossof analyte in any splitless injection, calculation of the necessary column capacity is not simple. Ifthe injection has been optimized for splitless hold-time and fused silica wool is being used in theliner to minimize high molecular weight discrimination, then it is easier to overload the analyticalcolumn. Possibly the biggest cause of overload is from pressure-pulsing the injection port, as thisimproves the transfer of all compounds to the column. The required capacity for your system willbe a function of the specific calibration standards and, more importantly, the injection port.
From a capacity consideration, a 0.25mm ID column with 0.25µm film thickness does nothave sufficient capacity for a 160ng per component standard. Figure 4a shows the poor peakshape observed when a column is overloaded. Increased capacity can be achieved byincreasing column diameter or film thickness.
When increasing column diameter, the flow rate of the column can be a concern with bench-topGC/MS systems. Many bench-top GC/MS systems do not have the pumping capacity for thecarrier gas flow that is needed with a 0.32mm ID column. A 0.28mm ID column can increasesample capacity without exceeding the pumping capacity of most bench-top GC/MS systems,making it ideal for calibrating semivolatile compounds from 20 to 160ng without overload.Alternatively, a 0.25mm ID column with a 0.5µm film thickness also has sufficient capacity tohandle a calibration from 20 to 160ng without exhibiting overload. Figure 4b shows excellentpeak shape for a 160ng-per-component standard on a 30m, 0.25mm ID, 0.5µm Rtx®-5Sil MScolumn.
The total analysis time should be as short as possible without sacrificing separation orresolution between compounds with similar mass spectra. Pay particular attention to theseparation between benzo-b- and benzo-k-fluoranthrene—they tend to be the most difficult-
Figure 4a & 4b. Avoid overload by selecting a column with the proper capacity.
1
2
3
4
5
6,7
8
1
2
3
4
5
6,7
8
Peak List for Fig. 4a & 4b1. di-n-octyl phthalate2. benzo(b)fluoranthene3. benzo(k)fluoranthene4. benzo(a)pyrene5. perylene-d126. indeno(1,2,3-cd)pyrene7. dibenz(a,h)anthracene8. benzo(ghi)perylene
4aPoor peak shape for a0.25mm ID, 0.25µm
Rtx®-5Sil MS column
4bExcellent peak shape for a
0.25mm ID, 0.5µmRtx®-5Sil MS column
to-separate analytes, sharing common mass spectra and quantitation ions. Figure 5 shows a80ng per component injection of the compounds listed in Table I with an analysis time under30 minutes. The expanded sections of the chromatogram show the excellent resolution thatcan be achieved with the Rtx®-5Sil MS column.
In the past, the GC/MS systems used for semivolatile analysis did not have the sensitivity forsplit injections, so laboratories were limited to splitless injection. Newer systems such as theAgilent 5973 and ion trap GC/MS have greatly improved sensitivity, which allow the use ofsplit injection and still meet the detection limits required by most semivolatile methods.Figure 6 shows the 20ng per component standard injected in split mode using a 20:1 split ratioon a 30m, 0.25mm ID, 0.25µm Rtx®-5Sil MS column. The low bleed exhibited by thiscolumn is critical when working with these more sensitive GC/MS systems. A benefit of splitinjection is narrower peak widths for improved separations between closely eluting com-pounds. Also, split injections usually result in less reactive compound breakdown because theresidence time in the injection port is much shorter than in splitless injection.
If the sensitivity of an instrument allows for split injection, then column capacity is not nearlythe issue it is for splitless injection. Figure 7 shows a 160ng-per-component standard injectedunder the same conditions as shown in Figure 6. A column with a thinner film can be usedbecause the concentration reaching the column is reduced by 20-fold. The analytical systemusing split injection will be able to handle higher concentrations of contaminants and possiblystay calibrated longer, but there will be a sacrifice in method detection limits (MDLs).Therefore, it is important to ensure that the MDLs specified in a particular method still can bemet if split injection is used.
Reducing DiscriminationReduced response of the higher molecular weight semivolatile compounds can be caused bydiscrimination in the injection port. In extreme cases response for the last three PAH com-pounds may be lost completely at lower calibration levels. To reduce the effects of discrimina-tion in the injection port, we recommend using a drilled Uniliner® inlet liner. Because thecolumn seals into the taper at the bottom of the liner, there is reduced loss of high molecularweight compounds and improved response. The drilled Uniliner® inlet liner has a small holedrilled at the entrance that allows it to work with small diameter columns and ElectronicPressure Control (EPC) injection systems. When using the drilled Uniliner® liner (Figure 8),the response of the last three PAHs is significantly higher compared to the same analysis donewith a normal splitless sleeve (see Figure 5). The drilled Uniliner® is available for Agilent5890 and 6890 GCs (see page 19).
Figure 8. The Rtx®-5Sil MS combined with the Siltek™ drilled Uniliner® liner exhibits excellent peak shape and response forthe semivolatile compounds listed in US EPA Method 8270.
20°C min. (hold 0 min.), to 330° @6°C/min. (hold 1 min.)
Det. type: MSTransfer line temp.: 280°CScan range: 35 to 550amuIonization: EIMode: full scan
GC_EV00576
16
QuantitationBecause splitless injections suffer from irreproducibility, quantitation for semivolatilecompounds is by internal standard, using a single ion for each analyte. Internal standards atknown concentrations are used to correct for variances in the amount of material transferredto the column with different injections, and also to track MS sensitivity.
All sample analyte concentrations are calculated using response factors obtained from thecalibration curve. This method uses single ions (i.e., extracted ions) for each compound sothat chromatographic resolution of each compound is not necessary. It is acceptable forcompounds to co-elute, as long as they do not have any common ions used for quantitation.Typically, only isomers have similar spectra, so chromatographic resolution is required onlyfor these compounds.
Figure 9 shows the possibilities for calibration curves for this analysis. Curve A is the desiredresult, indicating a proportional response with increasing concentration. This implies thatthere is no detector saturation or reactivity for this compound. Curve B is indicative of acompound that undergoes a reaction, usually in the injection port or on the head of thecolumn. If a calibration curve like Curve B is observed, injection port or column maintenanceis required. Finally, Curve C shows high-end roll-off, indicating saturation. If a calibration
Figure 9. Calibration curves showing linear response, adsorption, and high-end roll-off.
resp
once
(are
a co
unts
)
concentration
CA
B
20000
40000
60000
80000
100000
120000
140000
160000
180000
200000
20 40 90 80 100 120 140 160
curve like Curve C is observed, select a column with higher capacity or reduce detectorsensitivity by lowering the multiplier voltage.
SummaryAlthough the analysis of semivolatile organic compounds is one of the more difficult testsperformed by environmental laboratories, using Restek’s Rtx®-5Sil MS or Rtx®-5MS columnsand following the tips presented in this guide can make it easier to perform. Correct samplepreparation, extract cleanup, injection technique, analytical columns, standards, and quantitationcan help minimize problems normally associated with semivolatile organic analyses. Whenproblems occur, use the most appropriate troubleshooting and maintenance procedures to quicklyoptimize your analytical system. When faced with difficulties in your semivolatile analysis,remember that the majority of problems occur in the sample preparation and cleanup step, or inthe GC injection port. If you still are having difficulties after reading through this guide, contactRestek’s Technical Service Team via email at [email protected] or via phone at800-356-1688 or 814-353-1300, ext. 4.
Rtx®-5MS and Rtx®-5Sil MS ColumnsConventional capillary gas chromatography (GC) columns use liquid stationary phases, manyof which are crossbonded to yield a higher working temperature. Even with crossbonding,however, the liquid stationary phase will slowly elute. This elution of the stationary phase,also termed column bleed, is more detectable at higher temperatures and is typically observedas an increasing baseline that follows the oven temperature program. Depending on themethod of detection, column bleed may not be an issue for certain separations. If the capillarycolumn is connected to a sensitive detector like a mass spectrometer (MS), then column bleedcan cause a number of problems—specifically misidentification of analytes, loss of sensitiv-ity, and inaccurate quantitation.
The level of column bleed will affect the senstivity of any MS, especially ion trap instru-ments, which use automatic gain control. As the level of column bleed increases, so does thesignal from bleed ions in the mass spectra of analytes and unknowns. Also, sensitivity (ordetection limit) severely degrades. The contribution of bleed ions to the mass spectra canresult in misidentification of compounds, requiring laboratory personnel to subtract these ionsbefore performing library searches. Doing so can add considerable time to their analyses.Finally, if bleed ions contribute to the signal of the quantitation mass, quantitation of analytesand unknowns will be miscalculated. For these reasons, it is critical that analysts choose thelowest-bleed column designed for GC/MS applications.
Many manufacturers offer “MS” phases for applications requiring low bleed. In many cases,these represent nothing more than the reporting of the bleed signal when the column wastested for a single analysis at the manufacturer. Restek has developed true low-bleed MSphases. These columns exhibit a much lower column bleed than was previously available. TheRtx®-5MS column is a low-bleed, dimethyl/diphenyl polysiloxane phase. The Rtx®-5Sil MS isa low-bleed silarylene methyl/phenyl phase. The combination of Restek’s polymer chemistryand rigorous QA testing ensures that each MS column exceeds requirements of the mostsensitive mass spectrometers.
Rtx®-5Sil MSFused Silica (equivalent selectivity of Crossbond®
5% diphenyl/95% dimethyl polysiloxane)Stable to 360°C
ID df (µm) temp. limits 15-Meter 30-Meter 60-Meter0.25mm 0.10 -60 to 330/350°C 12605 12608 12611
0.25 -60 to 330/350°C 12620 12623 126260.50 -60 to 330/350°C 12635 12638 126411.00 -60 to 325/350°C 12650 12653
0.32mm 0.10 -60 to 330/350°C 12606 12609 126120.25 -60 to 330/350°C 12621 12624 126270.50 -60 to 330/350°C 12636 12639 126421.00 -60 to 325/350°C 12651 12654
0.53mm 0.50 -60 to 320/340°C 12637 126401.00 -60 to 320/340°C 12652 126551.50 -60 to 310/330°C 12667 12670
ID df (µm) temp. limits 15-Meter 30-Meter0.25mm 0.10 -60 to 330/350°C 12705 12708
0.25 -60 to 330/350°C 12720 127230.50 -60 to 330/350°C 12735 127381.00 -60 to 325/350°C 12750 12753
0.28mm 0.25 -60 to 330/350°C 12790 127930.50 -60 to 330/350°C 12791 127941.00 -60 to 325/350°C 12792 12795
0.32mm 0.10 -60 to 330/350°C 12706 127090.25 -60 to 330/350°C 12721 127240.50 -60 to 330/350°C 12736 127391.00 -60 to 325/350°C 12751 12754
0.53mm 0.50 -60 to 320/340°C 12737 127401.00 -60 to 320/340°C 12752 127551.50 -60 to 310/330°C 12767 12770
18
For detailed information on typesof guard columns, their uses, and
a complete product listing,request Restek’s
Guard Column Fast Facts flyer(lit. cat.# 59319)
Intermediate-Polarity Deactivated Guard ColumnsNominal ID (mm) Nominal OD (mm) 1-Meter 5-Meter
0.15 0.363 ± 0.012 10101 10042
0.18 0.37 ± 0.04 10102 10046
0.25 0.37 ± 0.04 — 10043
0.28 0.37 ± 0.04 — 10003
0.32 0.45 ± 0.04 — 10044
0.45 0.69 ± 0.05 — 10005
0.53 0.69 ± 0.05 — 10045
formoreinfo
Innovative Integra-Guard™ ColumnsSome people swear by press-fit connectors, and others swear at them. For many analysts, theart of attaching a guard column to an analytical column is a mystery. Restek’s chemists havediscovered the solution to this mystery—the easiest, most reliable connection is no connectionat all! No guard column system is more permanent than one continuous length of tubingcontaining both the guard column and the analytical column.
Restek offers a wide variety of Integra-Guard™ capillary columns with a guaranteed leak-freeconnection! The guard column is tied separately from the analytical column, using high-temperature string. The transition area between the columns is the point at which the guardcolumn ends and the analytical column begins. The entire setup is suspended in our unique“crush-free” cage, which prevents the column from coming in contact with anything thatcould damage it.
Ordering is simple. Just add the appropriate suffix number and price to the analytical column’scatalog number and price. For example, a 30m, 0.25mm ID, 0.25µm Rtx®-5MS with a 5mIntegra-Guard™ column is cat.# 12623-124.
ID (mm) Length (m) Suffix #0.25 5 -124
10 -127
0.28 5 -243
10 -244
Integra-Guard™
built-in guardcolumn
Phases currently available asIntegra-Guard™ columns:
For analysts who prefer to attach a guard column to the analytical columnthemselves, Restek offers deactivated guard columns and Press-Tight® connectors.
Press-Tight® Connectors• Seals all common sizes (0.18 to 0.53mm ID, outside diameters from 0.3 to 0.75mm) of
fused silica tubing.• Connect guard columns to analytical columns, repair broken columns, or connect column
outlets to transfer lines.• Angled connectors are designed to approximate the curvature of a capillary column and reduce
strain on column-end connections.• Made from inert fused silica.
Splitless Liners for ID*/OD & cat.# Similar toAgilent/HP & Finnigan GCs Benefits/Uses: Length (mm) ea. 5-pk. 25-pk. Agilent part #
trace samples >2µL 4.0 ID 20772 20773 20774 —4mm Splitless 6.5 OD x 78.5
trace samples >2µL 4.0 ID 22400 22401 22402 19251-605404mm Splitless w/ FS Wool 6.5 OD x 78.5
trace samples >2µL 4.0 ID 20912 20913 — —4mm Splitless (quartz) 6.5 OD x 78.5
trace samples >2µL 4.0 ID 22403 22404 — 18740-802204mm Splitless (quartz) w/ FS Wool 6.5 OD x 78.5 5181-8818
trace samples <2µL 2.0 ID 20795 20796 20797 —Gooseneck Splitless (2mm) 6.5 OD x 78.5
trace samples >2µL 4.0 ID 20798 20799 20800 5181-3316Gooseneck Splitless (4mm) 6.5 OD x 78.5
trace samples >2µL 4.0 ID 22405 22406 22407 5062-3587 Gooseneck Splitless (4mm) w/ FS Wool 6.5 OD x 78.5
trace, active samples >2µL 4.0 ID 20784 20785 20786 5181-3315Double Gooseneck Splitless (4mm) 6.5 OD x 78.5
trace, active, dirty 2.0 ID 20907 20908 — —Cyclo Double Gooseneck (2mm) samples <2µL 6.5 OD x 78.5
trace, active, dirty 4.0 ID 20895 20896 20997 —Cyclo Double Gooseneck (4mm) samples >2µL 6.5 OD x 78.5
allows direct injection when 4.0 ID 21054 21055 — —Drilled Uniliner® (4mm) using EPC-equipped GC 6.3 OD x 78.5
allows direct injection when 4.0 ID 21054-214.1 21055-214.5 — —Siltek™ Drilled Uniliner® (4mm) using EPC-equipped GC 6.3 OD x 78.5C
OL
UM
NI
NS
TA
LL
ST
HI
SE
ND
Inlet LinersFor Agilent/HP & Finnigan GCs
Splitless Liners for ID**/OD & cat.# Similar toVarian 1075/1077GCs Benefits/Uses: Length (mm) ea. 5-pk. 25-pk. Varian Part #
trace samples<2µL 2.0 ID 20721 20722 20723 01-900109-052mm Splitless 6.3 OD x 74
trace samples>2µL 4.0 ID 20904 20905 20906 01-900109-054mm Splitless 6.3 OD x 74
trace, active samples 4.0 ID 20847 20848 20849 —Double Gooseneck up to 4µL 6.3 OD x 74
trace, dirty, active 4.0 ID 20897 20898 — —Cyclo Double Gooseneck samples up to 4µL 6.3 OD x 74
1078/1079 Liners ID**/OD & cat.# Similar tofor Varian GCs Benefits/Uses: length (mm) ea. 5-pk. 25-pk. Varian Part #
trace samples <2µL 2.0 ID 21711 21712 — 03-918466-001078/1079 Splitless 5.0 OD x 54C
OL
UM
N I
NS
TA
LL
S T
HIS
EN
D
*Nominal ID at syringe needle expulsion point.
20
Splitless Liners for ID**/OD & cat.# Similar toPerkinElmer GCs Benefits/Uses: Length (mm) ea. 5-pk. 25-pk. PE Part #
trace samples 2.0 ID 20829 20830 20831 N6101372Auto SYS Splitless w/Wool (2mm ID)* 6.2 OD x 92.1
trace, active samples 3.5 ID 20853 20854 — —Auto SYS Double Gooseneck up to 4µL 6.2 OD x 92.1
trace, dirty, active samples 3.5 ID 20899 20900 — —Auto SYS Cyclo Double Gooseneck up to 4µL 6.2 OD x 92.1
most common 2.0 ID 21717 21718 — N612-1004Auto SYS XL Split/Splitless analyses 4.0 OD x 86.2
Splitless Liners for ID**/OD & cat.# Similar to5000-6000 Series GCs Benefits/Uses: Length (mm) ea. 5-pk. 25-pk. TO Part #
trace samples 4.0 ID 20814 20815 20816 —Splitless (4mm ID) 5.5 OD x 79.5
Splitless Liners for 8000 ID**/OD & cat.# Similar to& TRACE™ Series GCs Benefits/Uses: Length (mm) ea. 5-pk. 25-pk. TO Part #
Splitless Liners for ID**/OD & cat.# Similar toShimadzu GCs Benefits/Uses: Length (mm) ea. 5-pk. 25-pk. Shimadzu Part #
trace samples 3.5 ID 20955 20956 20957 221-0914594mm Splitless with Wool* 5.0 OD x 94
reduces backflash and 3.5 ID 20958 20959 20960 —94mm Double Gooseneck catalytic decomposition 5.0 OD x 94
reduces backflash, also 3.5 ID 20961 20962 20963 221-41599-0094mm Single Gooseneck operates in DI mode 5.0 OD x 94
CO
LU
MN
IN
ST
AL
LS
TH
IS E
ND
trace samples 3.0 ID 20942 20943 20944 453 20032Splitless (3mm ID) 8.0 OD x 105
trace samples 5.0 ID 20945 20946 20947 453 20033Splitless (5mm ID) 8.0 OD x 105
trace, active samples 4.0 ID 20952 20953 — —Double Gooseneck up to 4µL 8.0 OD x 105
Inlet LinersFor Shimadzu GCs
For PerkinElmer GCs
For Thermo Orion GCs
*Liner is packed with fused silica wool. To order glass wool instead, add the suffix “–202” to the liner’s catalog number.**Nominal ID at syringe needle expulsion point.
CO
LU
MN
IN
ST
AL
LS
T
HIS
E
ND
O-RingsViton®
Viton® o-rings are universal. One size fits both split (6.3mm ID) and splitless (6.5mm ID) sleeves.
Max. temp. Similar to Agilent Part # cat.# Qty.Viton® (fluorocarbon) 350°C 5180-4182 20377 25-pk.
GraphiteGraphite o-rings have excellent thermal stability and can be used at injection port temperatures up to 450°C!
Similar to Agilent Part # 10-pk. 50-pk.6.35mm ID for split liners 5180-4168 20296 20297
6.5mm ID for splitless liners 5180-4173 20298 20299
21
Deactivated Fused Silica Wool• Ensure uniform vaporization in split or splitless liners.• Prolong column life by trapping septum particles.• Recommended for autosamplers with fast injection rates.• Inertness tested for endrin breakdown.cat.# 20790, (10 grams)
Thermolite® Septa• Each batch tested on FIDs, ECDs, & MSDs to ensure lowest bleed.• Excellent puncturability.• Preconditioned and ready to use.• Do not adhere to hot metal surfaces.• Usable to 340°C inlet temperatures.• Packaged in non-contaminating tins.
Call our literature hotlineat 800-356-1688
or 814-353-1300, ext. 5,or your local Restek
representative forRestek’s 20-page bulletin,
A Guide to MinimizingSepta Problems
(lit. cat.# 59886).
guidefree
Replacement Inlet Seals• Special grade of stainless steel deforms easily, ensuring a completely leak-free seal.• Available in stainless steel, gold-plated, and Silcosteel®-treated.• Cross-Disk ideal for high-flow split applications on EPC-equipped GCs.• Shipped with washers.
*0.8mm ID stainless steel inlet seal is equivalent toAgilent part #18740-20880.**0.8mm ID gold-plated inlet seal is equivalent toAgilent part #18740-20885.
Single-Column Installation,Opening Size 0.8mm ID
Stainless Steel Inlet Seal*
21315, 2-pk. 21316, 10-pk.
Gold-Plated Inlet Seal**
21317, 2-pk. 21318, 10-pk.
Silcosteel® Inlet Seal
21319, 2-pk. 21320, 10-pk.
For Agilent 5890/6890/6850Split/Splitless Injection Ports Cross-Disk, Opening Size 0.8mm ID
Gold-Plated Inlet Seal
20477, 2-pk. 20476, 10-pk.
Silcosteel® Inlet Seal
20475, 2-pk. 20474, 10-pk.
Cross-Disk, Opening Size 1.2mm ID
Gold-Plated Inlet Seal
21009, 2-pk. 21010, 10-pk.
Silcosteel® Inlet Seal
21011, 2-pk. 21012, 10-pk.
Cross-Disk for Agilent GCs†
†Similar to Agilent part #5182-9652.
Request the handy,pocket-sized, InletSupplies Guide(lit. cat.# 59893A).
Ferrule Fits Column Vespel®/ID (mm) ID (mm) Qty. Graphite Graphite
0.4 0.25 50-pk. 20227 20229
0.4 0.25 10-pk. 20200 20211
0.5 0.32 10-pk. 20201 20212
0.5 0.32 50-pk. 20228 20231
0.6 0.28 10-pk. — 20232
0.8 0.53 10-pk. 20202 20213
0.8 0.53 50-pk. 20224 20230
Ferrules
Compact Ferrules for Agilent GCs
Ferrule Fits Column Vespel®/ID (mm) ID (mm) Qty. Graphite Graphite
0.4 0.25 10-pk. 20250 20238
0.4 0.25 50-pk. 20251 20239
0.5 0.32 10-pk. 21007 20248
0.5 0.32 50-pk. 21008 20249
0.8 0.53 10-pk. 20252 20263
0.8 0.53 50-pk. 20253 20264
Encapsulated Ferrules• Will not deform and stick in fittings.• Reusable.• For 1/16" compression fittings.
Ferrule ID Fits column ID cat.#/10-pk.0.4mm 0.25mm 21036
0.5mm 0.32mm 21037
0.8mm 0.53mm 21038
EZ-Vent™ 2000• Change GC/MS columns in minutes without venting.• Silcosteel® treated for greater inertness.• Deactivated transfer line minimizes bleed into the source.• Stainless steel body and high-temperature polyimide
ferrules minimize leaks at the problematic transfer line fitting.• Less expensive than other “no-vent” fittings.• 100µm transfer line throttles vacuum and prevents column pump down.• Available for Agilent GCs with 5971/5972 or 5973 MS and Varian 3400, 3600, or 3800
GCs with Saturn 2000 MS.• Precision-machined orifice.
KitsEZ-Vent™ 2000 for Agilent GCs with 5971/5972 or 5973 MS
Includes EZ-Vent™ 2000, 1/16" SS nut, 0.4mm ID ferrules for connecting capillary column,0.4mm ID ferrules for connecting transfer line, 100µm deactivated transfer line (3 ft.), andEZ-Vent™ column plug; cat.# 21013, (kit)
EZ-Vent™ 2000 for Varian Saturn 2000 MS systems with 3400, 3600, or 3800 GCsIncludes EZ-Vent™ 2000, 1/16" SS nut, 0.4mm ferrules for connecting capillary column,0.4mm ID ferrules for connecting transfer line, 100µm deactivated transfer line (3 ft.), andEZ-Vent™ column plug; cat.# 21014, (kit)
great
idea!EZ-Vent™ 2000 for Agilent GCs
EZ-Vent™ 2000 for Varian GCs
EZ-Vent™ 2000 ferrules
23
For Agilent 5890 GCsDescription cat.#, (ea.)
Replacement Weldment (Similar to Agilent part# 19251-60575) 20265
Replacement Shell Weldment (Similar to Agilent part# 19251-80570) 20266
Silcosteel® Weldment 20267
Silcosteel® Shell Weldment 20268
Direct Replacement Split/Splitless Injection Ports for Agilent GCs
Would you like better performance from your injector? Restek’s Silcosteel®-coated split/splitless injector is a direct replacement for Agilent 5890 and 6890/6850 GCs. The injector
version available
®
For Agilent 6890/6850 GCsDescription cat.#, (ea.)
Replacement Weldment for Agilent 6890/6850 GCs with EPC 22674
Replacement Weldment for Agilent 6890/6850 GCs with manual flow 20265
Replacement Shell Weldment for Agilent 6890/6850 GCs 22673
Weldment for Agilent 5890 GCs
MSD Source NutThe nut bore has been changed from 0.8mm to 1.2mm to permit easy removal of ferrules witha standard tapered-needle file (cat.# 20106). The nuts still match the manufacturer’s originalpart specifications and are made of brass to prevent thread-stripping on the transfer line.(Similar to Agilent part #05988-20066.)(Detector) MSD Source Nut: cat.# 20643, (2-pk.)
Agilent’s MSD interfacerequires a butt-seal at thebase of a Vespel® ferrule,
which is prone to leakage.Restek’s version uses a
standard ferrule design thatsimultaneously seals the
fitting and capillary tubingwith compressive forces.
MSD Conversion Fitting—Improved• Uses a flat, soft aluminum sealing ring to deform and butt-seal against the MSD interface
(see figure below).• A standard Vespel® ferrule seals the column and 1/16-inch stainless steel nut.• Fitting is constructed of nickel-plated brass for longevity and softness.• Can use any standard Vespel® or Vespel®/graphite 1/16-inch ferrule.• Includes a 1/16-inch stainless steel nut and two replacement sealing rings. Order ferrules
is manufactured from high-quality stainless steel andmeets or exceeds Agilent original equipment specifica-tions. Silcosteel® passivates the metal surface to ensurean inert pathway for the sample, delivering increasedperformance.
Shell weldment for Agilent5890 GCs
Weldment and shellweldment for Agilent
6890/6850 GCs
✓ check it out
24
Internal Standard Mixtures
SV Internal Standard Mixacenaphthene-d10 naphthalene-d8crysene-d12 perylene-d121,4-dichlorobenzene-d4 phenanthrene-d10
04.1 SOW, 04.2 OSWRestek chemists carefully reviewed the 04.2 Statement of Work and determined that theidentical products listed in 04.1 will also be required for the 04.2 revision. The products listedhere are a result of this work.
CLP 04.1 B/N MegaMix™
Note: This product is provided as a two ampul set:
Qualitative mixture useful for determining GPCdump/collect times. Data packs are not available.The compounds are dissolved in methylenechloride at the concentrations listed.
Benzaldehyde and atrazine will reactquickly and directly with the methanolstabilizer used in most brands andgrades of methylene chloride. Thisreaction will prevent you fromobtaining stable, working-levelcalibration standards. Therefore, Restekhas prepared the CLP OLM 04.1Semivolatile B/N Mega- Mix™ frommethylene chloride that is stabilizedwith amylene and is completely free ofmethanol. Restek strongly recommendsscreening the methylene chloride youuse to dilute these mixtures to confirmthat it is free of methanol.
3,3'-Dichlorobenzidine2,000µg/mL each in methanol, 1mL/ampul
Each 5-pk. 10-pk.31026 31026-510
w/data pack 31026-500 31026-520 31126
PATENTS & TRADEMARKSRestek patents and trademarksare the property of RestekCorporation. Other trademarksappearing in Restek literatureor on its website are the prop-erty of their respective owners.